A. Using MIMs in coordination with gene expression datasets to unravel cell regulatory networks.This project is based on the premise that genes that are co-regulated in a variety of cell types are likely to be functionally related. We developed that concept utilizing data mining software developed in our Laboratory (?CellMiner?) to analyze gene expression data in the NCI-60 human tumor cell lines.We have also now published the CellMiner software tools and made them generally available:Reinhold, W.C., Sunshine, M., Liu, H., Varma, S., Kohn, K.W., Morris, J., Doroshow, J., and Pommier, Y. (2012). CellMiner: A Web-Based Suite of Genomic and Pharmacologic Tools to Explore Transcript and Drug Patterns in the NCI-60 Cell Line Set. Cancer Res 72, 3499-3511.Utilizing those facilities, we investigated control networks involved in cell migration/invasion and molecular interactions specific to cells having epithelial character, as follows.A1. Molecular interaction network model of cell migration/invasion processes.We began with a set of 70 genes derived by a new clustering procedure designed by our collaborator, Barry Zeeberg (see reference below) that links expression in the NCI-60 panel of human tumor cells with functions assigned by the Gene Ontology (GO) database.Zeeberg, B.R., Reinhold, W., Snajder, R., Thallinger, G.G., Weinstein, J.N., Kohn, K.W., and Pommier, Y. (2012). Functional categories associated with clusters of genes that are co-expressed across the NCI-60 cancer cell lines. PLoS One 7, e30317.After adding other NCI-60 expression-correlated genes to one of Zeeberg?s gene clusers, we derived a set of 32 genes whose expressions were mutually correlated with each other. We found that most of those genes had functions related to interactions with extracellular matrix. Utilizing molecular interaction maps (MIMs), we published a coherent picture of this regulatory network: Kohn, K.W., Zeeberg, B.R., Reinhold, W.C., Sunshine, M., Luna, A., and Pommier, Y. (2012). Gene expression profiles of the NCI-60 human tumor cell lines define molecular interaction networks governing cell migration processes. PLoS One 7, e35716.A2. Epithelial gene expression signatures from the NCI-60 human tumor cell lines.In a new approach to the characterization of cells having epithelial character, we focused on the expression of genes for the proteins involved in tight-junctions, structures that are required to maintain polarity of epithelial cells. From our NCI-60 gene expression data, we identified a subset of 12 cell lines whose selective expression of a given gene serves as a consensus pattern or signature for that gene having specific functions in tumor cells of epithelial character. We collected a 75 genes that most closely fit that consensus pattern and investigated their known functions from evidence in the scientific literature. Most of the genes had known epithelial functions, although many did not, usually due to very little published information about them. Since the latter genes had expression patterns in the NCI-60 that closely fit the epithelial consensus, their further investigation from that point of view is likely to be fruitful. We are preparing molecular interaction maps to provide an integrated view of the functional features characteristic of epithelial-type cancer cells. This will help to distinguish cancer cell types amenable to different therapies.B. Enhancing the utility of our Molecular Interaction Map (MIM) notation. In order to make the MIM notation readily accessible to the scientific community and make it easier to learn, as well as to provide tools for developers of MIM applications, we have developed the following software:B1. PathVisio-MIMWe developed and published a plug-in to PathVisio for creating and editing MIMs and we now use this tool for creating and editing MIMs:Luna, A., Sunshine, M.L., van Iersel, M.P., Aladjem, M.I., and Kohn, K.W. (2011). PathVisio-MIM: PathVisio plugin for creating and editing Molecular Interaction Maps (MIMs). Bioinformatics 27, 2165-2166. B2. Tools for MIM developersIn order to provide tools for MIM software development based on our MIM notation rules (Kohn et al., Mol. Biol. Cell 17: 1-13, 2006), we published the following Research Article:Luna, A., Karac, E.I., Sunshine, M., Chang, L., Nussinov, R., Aladjem, M.I., and Kohn, K.W. (2011). A formal MIM specification and tools for the common exchange of MIM diagrams: an XML-Based format, an API, and a validation method. BMC Bioinformatics 12:167.B3. PathVisio ValidatorWe collaborated with developers of PathVisio and SBGN (Systems Biology Graphics Notation) to develop a validation framework in PathVisio that can be used with rule sets for MIM or other diagram notations, and we published the following Systems Biology note:Chandan, K., van Iersel, M.P., Aladjem, M.I., Kohn, K.W., and Luna, A. (2012). PathVisio-Validator: a rule-based validation plugin for graphical pathway notations. Bioinformatics 28, 889-890.The Validator will help users check whether their diagrams conform to a standard rule set, such as for MIM, so as to make the diagrams unambiguous and coded in a manner suitable for exchange with other users. C. Functions of MdmX in Fine-Tuning the Response of p53 to DNA DamageThe regulation of the response of p53 to DNA damage is key to the ability of cancer cells to survive following treatment with cytotoxic drugs or radiation. In cells having the normal functioning gene, p53 induced by such treatments promotes the cell?s DNA repair capabilities and causes cell cycle delays to allow more time for repair to take place before the cell begins mitosis, the critical cell division event. The p53 induction however must be tightly controlled to avoid overshooting p53 production, which can cause cells to die by apoptosis. Understanding how this control works is important for designing therapy to protect normal cells while selectively killing cancer cells. We think that one mechanism for this control involves the gene MdmX, also known as MDM4. MdmX interacts with and alters the activites of both p53 and MDM2, the major inhibitor and destroyer of p53. Therefore we are investigating the role of MdmX in the p53 response to DNA damage.Based on a molecular interaction map (MIM) of the p53-Mdm2-MdmX control network, we simulated a network model and surveyed parameter space to disclose regions exhibiting different types of p53 responses:Kim S, Aladjem MI, McFadden GB, Kohn KW. Predicted functions of MdmX in fine-tuning the response of p53 to DNA damage. PLoS Comput Biol. 6(2):e1000665, 2010.A major finding was that MdmX/MDM4 can dampen the p53 oscillations that occur after DNA damage. That action may prevent high transient p53 activity at could lead to inappropriate cell death. We have now engineered cell lines to test this hypothesis at the single-cell level and more generally to investigate the role of MdmX in the response of p53 to ionizing radiation-induced DNA damage. We measure p53 gene activity by assay of the mRNA in living cells using a bacteriophage MS2 system. The induced p53 gene was engineered to have an added sequence of several MS2 coat protein-binding stem-loops in the mRNA. The MS2 coat protein binding to this mRNA is measure by fluorescence of a GFP adduct on the transduced MS2 coat protein. We have successfully implemented this state-of-the-art system and are now working to optimize the signal-to-noise ratio.